• invasiveness;
  • light acclimation;
  • light harvesting;
  • nitrogen use-efficiency;
  • phenotypic variability;
  • plasticity;
  • support costs


  1. Top of page

To understand the role of leaf-level plasticity and variability in species invasiveness, foliar characteristics were studied in relation to seasonal average integrated quantum flux density (Qint) in the understorey evergreen species Rhododendron ponticum and Ilex aquifolium at two sites. A native relict population of R. ponticum was sampled in southern Spain (Mediterranean climate), while an invasive alien population was investigated in Belgium (temperate maritime climate). Ilex aquifolium was native at both sites. Both species exhibited a significant plastic response to Qint in leaf dry mass per unit area, thickness, photosynthetic potentials, and chlorophyll contents at the two sites. However, R. ponticum exhibited a higher photosynthetic nitrogen use efficiency and larger investment of nitrogen in chlorophyll than I. aquifolium. Since leaf nitrogen (N) contents per unit dry mass were lower in R. ponticum, this species formed a larger foliar area with equal photosynthetic potential and light-harvesting efficiency compared with I. aquifolium. The foliage of R. ponticum was mechanically more resistant with larger density in the Belgian site than in the Spanish site. Mean leaf-level phenotypic plasticity was larger in the Belgian population of R. ponticum than in the Spanish population of this species and the two populations of I. aquifolium. We suggest that large fractional investments of foliar N in photosynthetic function coupled with a relatively large mean, leaf-level phenotypic plasticity may provide the primary explanation for the invasive nature and superior performance of R. ponticum at the Belgian site. With alleviation of water limitations from Mediterranean to temperate maritime climates, the invasiveness of R. ponticum may also be enhanced by the increased foliage mechanical resistance observed in the alien populations.


  1. Top of page

Many introduced plant species have become particularly competitive and invasive in new environments (Mooney 1999), constituting a danger to local biodiversity. Although such biological invasions have become increasingly frequent, plant ecophysiological and ecological traits underlying these invasions are still poorly understood. Global change phenomena such as elevated carbon dioxide concentrations as well as high nitrogen (N) deposition rates, and associated increases in the area of habitats that certain species can colonize, apparently provide one explanation for enhanced dispersal of plants that may be of limited competitive ability if the resource availability is limited (Dukes & Mooney 1999). High resource use efficiency as well as enhanced phenotypic plasticity have been hypothesized as decisive plant traits allowing the species to become invasive in newly colonized habitats (Abbott 1992; Dukes & Mooney 1999; Davis, Grime & Thompson 2000; Huxman & Smith 2001), but there are few ecophysiological comparisons of species growing in invaded and native stands as well as between native and invasive plant species (Yamashita et al. 2000; Durand & Goldstein 2001).

In the British Isles and in Belgium, Rhododendron ponticum L. is an alien, widely naturalized and very invasive species (Cross 1975; Rotherham 1983). It is considered as a threat to local biodiversity and as a forest weed that is difficult to control (Brown 1953; Tabbush & Williamson 1987). The primary native range of R. ponticum extends north and south to the Black Sea, in the Caucasus and northern Turkey. However, there are a few isolated populations in Lebanon, Portugal and southern Spain (Cross 1975). These isolated populations are relicts of Tertiary subtropical conditions in Europe, while most of the range of R. ponticum in the Mediterranean Basin has vanished with the onset of typical summer-dry climates (Palamarev 1989). Nowadays these populations are restricted to mountain ranges that mitigate the Mediterranean drought, and function as ecological refugiae for many Tertiary relicts (Hampe & Arroyo 2002). The isolated native populations of R. ponticum are currently declining due to land use changes (Madueño 1991; Blanca et al. 2000), and also possibly due to long-term changes in water availability. The plant has been included in the recent Red List of Threatened Vascular Plants of Andalucía (Blanca et al. 2000). According to latest genetic studies, the invasive populations of R. ponticum in the British Isles, and possibly also in Belgium primarily originate from the relict southern Spanish population (Milne & Abbott 2000).

The enhanced invasive potential of introduced individuals of R. ponticum in the British Isles and Belgium is still not entirely understood. High competitive potential of introduced R. ponticum has been related to its ability to tolerate shade while still maintaining large growth rates in high light, to its effective vegetative dispersal, as well as to its high concentrations of toxic protective compounds such as andromedotoxin and rhododendrin that essentially avoid herbivory of R. ponticum foliage (Cross 1975). It has been suggested that disturbance and selective grazing of non-toxic co-occurring species such as Ilex aquifolium L. are the major factors aiding to the invasion of this species (Cross 1981; Rotherham 1983).

The presence of defence compounds deters herbivores (Jela 1997; Jela & Lawton 1997), but it also brings about low litter decomposition rates in Rhododendron species (Cross 1975; Benfield & Webster 1991) due to antibacterial and antifungal activities of these protective chemicals (Walker et al. 1999; Reddy et al. 2001). Slow litter decomposition rates, in turn, may lead to limited nutrient availability for foliage construction. Thus, over a long term, high nitrogen immobilization in soil organic matter may potentially constrain the amount of light intercepted by the Rhododendron canopy, and allow establishment of seedlings of competing tree species. Given that the maximal height of R. ponticum canopy is approximately 4–6 m (Cross 1975), this may mean that long-term competitive success of R. ponticum is limited by low soil nitrogen availability, facilitating the establishment of overstorey tree seedlings. This may partly explain the non-invasiveness of this species in its native habitats. However, during the past few decades there has been a dramatic increase in N-deposition rates in many forest ecosystems. Such increases in N availability may allow leaf formation under lower irradiance levels, thereby leading to important increases in plant light-harvesting efficiency, and enhanced plant competitive potential in lower irradiances. The rise in N-deposition rates has also been paralleled by a sharp increase in the rate of Rhododendron invasion during the last 30 years (Tabbush & Williamson 1987). To gain insight into the enhanced N deposition loads on species-competitive potentials, we studied a native declining R. ponticum population from southern Spain, and an introduced very invasive population in Belgium. The atmospheric N deposition rates are low in the habitat of the native population, but high in the invasive range of R. ponticum. We asked whether the light-harvesting efficiencies differ between the various populations of R. ponticum, and whether part of the invasiveness of R. ponticum is associated with enhanced N availability in the western European forests.

There are indications that high phenotypic plasticity, and in particular, a highly plastic response to light, is involved in the invasive capacity of certain species (Yamashita et al. 2000; Durand & Goldstein 2001). To further understand the role of plasticity in species competitive potential, we have compared leaf-level plasticity in response to canopy light gradients between the native and the introduced populations of R. ponticum. We hypothesized that the leaf-level plasticity was higher in the Belgian population, allowing for a more efficient acclimation to various forest understorey and gap conditions in the alien population. As a contrast to the plastic response of R. ponticum, we included in the study the native species Ilex aquifolium, another evergreen, understorey shrub. The local abundance of I. aquifolium has dramatically decreased in Belgium and in the British Isles due to the greater competitiveness of R. ponticum, which has occupied the microsites of I. aquifolium. Even though there is detailed information on dispersal and reproduction of these two species in forest understoreys (Rotherham 1983; Obeso 1997; Esen 2000), comparative data of leaf- and canopy ecophysiology are scant. We hypothesized that the greater competitive ability of R. ponticum over I. aquifolium results from a more efficient N use in photosynthesis and an improved light-harvesting efficiency in low irradiance in R. ponticum, and that these species differences are specifically amplified by enhanced N deposition rates.


  1. Top of page

The Spanish site

The native population of Rhododendron ponticum L. was investigated in the Alcornocales Natural Park, Sierra del Aljibe, Cádiz province (Sierra del Cabrito, Llanos del Juncal area, 36°20′ N, 5°30′ E, elevation approximately 300–600 m). The study area supports Mediterranean sclerophyllous evergreen and semi-deciduous forests dominated by cork oak (Quercus suber L.), accompanied by Arbutus unedo L., Quercus canariensis Willd., Phillyrea latifolia L., Rhamnus alaternus L. and Erica arborea L. (Ojeda, Marañón & Arroyo 2000). These stands are 5–12 m tall, and have leaf area indices (LAI) of 2–5 m2 m−2. However, R. ponticum is primarily restricted to wetter patches in the understorey of the riparian gallery forests extending in narrow strips along small creeks (Aparicio, Porras & Ceballos 2001). In certain especially humid areas used for cattle grazing, it may also disperse onto open slopes, but in such locations is generally associated with springs. Shrubs or small trees of Ilex aquifolium are also restricted to more humid areas, which support vigorous understorey vegetation and stand LAI higher than 4 m2 m−2. Yet, I. aquifolium tolerates lower water availability than R. ponticum (authors’ observations), and may occasionally reach the overstorey to form mixed stands with A. unedo and Quercus spp.

The sandy soil is derived from Oligo-Miocenic sandstones, and is a dystric regosol according to FAO classification. The pH measured in the distilled water was 5.82 ± 0.14 (mean ± SE, n = 8) for the A horizon below the R. ponticum shrubs, and 5.22 ± 0.01 below the I. aquifolium shrubs (means were significantly different at P < 0.02 according to a t-test). Total N content of the A horizon was 0.49 ± 0.05% below R. ponticum, and 0.65 ± 0.09% below I. aquifolium, and C/N molar ratio was 21.9 ± 0.5 below R. ponticum, and 19.4 ± 0.6 below I. aquifolium (means were not significantly different for both comparisons, P > 0.1).

The climate is Mediterranean subhumid with average long-term yearly temperature of 16.7 °C (measured at Alcalá de los Gazules, approximately 10 km from the study site at elevation 221 m; Hampe & Arroyo 2002). The yearly average precipitation is 880 mm, and it is primarily concentrated in the autumn and winter months. Summer drought is alleviated by local fogs produced by the uplift of oceanic winds that are frequent in the area of the Strait of Gibraltar (Sinamba Difusión 1998). The moisture availability within the gorges of the Sierra del Aljibe is larger and seasonally less variable due to additional water supplied as fog throughout the year (Hampe & Arroyo 2002).

The Belgian site

The introduced invasive population of R. ponticum was studied in the understorey of an open Pinus sylvestris L. forest in Brasschaat, the province of Antwerpen (51°18′ N, 4°31′ E, elevation 16 m). The 73-year-old P. sylvestris plantation with a dominant height of 22–24 m and stocking of 556 trees ha−1 had an overstorey LAI of 1.8–2.4 m2 m−2 (Gond et al. 1999). The understorey woody layer was dominated by the exotic species Prunus serotina Ehrh. (height 3–9 m), and R. ponticum (height 1–5 m). Due to partial removal of the woody understorey, the shrub layer was non-uniformly distributed, such that LAI ranged from 1 to 5 m2 m−2 in different understorey patches (Nadezhdina, Tatarinov & Ceulemans 2003). In more open locations, R. ponticum often formed a thick monospecific canopy. In the denser parts of the understorey, taller trees of P. serotina provided extensive shading to shorter individuals of R. ponticum. Ilex aquifolium was present throughout the site, but generally as a minor understorey associate.

The stand is on a deep moderately wet Umbric Regosol (FAO classification) that has a distinct humus or iron B horizon. The litter layer (A0) thickness varies from 0 to 6 cm, being occasionally even thicker under the shrubs of R. ponticum. The pH determined in distilled water is moderately acidic (3.8–4.1) throughout the entire soil profile (Van Ranst et al. 2002). The area is characterized by high atmospheric nitrogen deposition rates of up to 30–45 kg N ha−1 year−1 (Neirynck, Roskams & Lust 1999; Neirynck et al. 2002), and the total N content of the litter layer is 1.11 ± 0.19%, and of the A horizon 0.91 ± 0.11% (C/N molar ratio 20.4 ± 0.7) (Janssens et al. 1999; Van Ranst et al. 2002). Foliar analyses indicate development of moderate soil cation and phosphorous limitations (Roskams, Sioen & Overloop 1997).

The temperate maritime climate is characterized by high yearly average precipitation of 767 mm that exceeds the yearly potential evaporation of 670 mm. According to 30-year records, the mean temperature is 3 °C for the coldest month and 18 °C for the warmest month, and the yearly average temperature is 9.8 °C (Janssens et al. 1999).

Foliage sampling for morphological and chemical analyses

Current-year foliage was sampled during 10–11 July 2002 at the Spanish site, and during 22–23 July at the Belgian site from a total of 42 individual plants of R. ponticum and I. aquifolium. Leaves were taken from the top to the bottom along the light gradient in the shrubs’ canopy. On each individual plant, one to ten canopy locations were selected, and four to six leaves from each canopy location were collected. Leaves were enclosed in plastic bags and maintained on ice at 0–5 °C until analysis in the laboratory on the next day. Sheets of wet filter paper were enclosed in the plastic bags with the leaves to avoid leaf transpiration and allow the leaves to reach full hydration (Gucci, Lombardini & Tattini 1997).

Leaf morphological measurements

For each leaf, lamina thickness (T) was measured in 5–14 (on average 8.1) leaf locations across both sides of the leaf mid-rib by digital precision callipers, and averages were calculated. Leaf area and leaf perimeter were measured using a photographic technique. Because the leaves were rarely completely flat, we measured both the horizontal projection of a leaf with its natural three-dimensional shape (silhouette area, AS), and after pressing the leaf to a flat section (projected area, AP). For photographing, the leaves were laid on a light table, and digital photographs were taken from a distance of 1.6 m using Nikon Coolpix 990 camera equipped with Tele Converter TC-E2 2x (Nikon Corporation, Tokyo, Japan). Objects of known area were photographed together with the leaves, and each image file was calibrated separately to obtain an appropriate pixel to centimetre conversion ratio. Repeated measurements of these calibration objects in different image files indicated that the error in area measurements was less than 1%. UTHSCSA Imagetool 2.00alpha (C. Donald Wilcox, S. Brent Dove, W. Doss McDavid and David B. Greer, Department of Dental Diagnostic Science, The University of Texas Health Science Center, San Antonio, TX, USA; was used to measure all estimates of foliar variables from the digital images.

Leaf laminas and petioles were separately weighed after oven drying at 70 °C for 48 h, and lamina (FL) and petiole (FP) dry to fresh mass ratios, and lamina dry mass per unit area (MA) were calculated. Lamina density (mass per unit volume) was determined as MA/T.

Seasonal integrated irradiance

Hemispherical photographs were taken above each leaf cluster using the Nikon Coolpix digital camera equipped with a Nikon Fisheye Converter FC-E8 0.21x. The hemispherical images were analysed with HemiView 2.1 (Delta-T Devices, Cambridge, UK), and the fractions of penetrating diffuse solar radiation of open sky (indirect site factor, ID) and of potential penetrating direct solar radiation of open sky (direct site factor, IB) were determined. Diffuse light distribution was calculated for uniformly overcast sky conditions. Calculations were carried out for eight azimuth classes and 20 zenith angle classes, giving 160 sky regions. IB was calculated separately for each month of the year. The mean IB for the months relevant for leaf development according to phenological observations at the study sites (March to June in Spain, April to July in Belgium) were calculated from the monthly values.

The relative irradiance indices determined from the hemispherical photographs were used to calculate average (1 March−30 June for the Spanish site, and 20 April−20 July for the Belgian site) daily integrated incident quantum flux densities during leaf growth and development (Qint, mol m−2 d−1). For the Spanish site, the hemispherical technique was calibrated using the monthly average values of above-canopy global solar radiation (RG, MJ m−2 d−1) measured in Cádiz (36°32′ N, 6°17′ E, Secretaría General Técnica, Ministerio de Medio Ambiente 2001):

Qint  =  γRG[IDpD  +  IB(1  −  pD)](1)

where γ (mol MJ−1) is the conversion factor for the global solar radiation, and pD is the proportion of diffuse quantum flux density (Niinemets & Kull 1998). Monthly values of γ were obtained from measurements in Almeria (36°83′ N, 2°41′ W, Alados, Foyo-Moreno & Alados-Arboledas 1996), yielding an average (March–June) above-canopy quantum flux density (γRG) of 40.0 mol m−2 d−1. An average value of pD of 0.263 was derived from the number of overcast and partly overcast days in Cádiz (Secretaría General Técnica, Ministerio de Medio Ambiente 2001) and measurements of atmospheric transmissivity and sky irradiance in Almeria (Alados-Arboledas et al. 2000).

At the Belgian site, the above-canopy quantum flux density was measured with a quantum sensor (190SA; Li-Cor, Inc., Lincoln, NE, USA), and half-hour averages were stored. These measurements yielded an average (20 April−20 July) above-canopy Qint of 29.9 mol m−2 d−1. A pD of 0.513 was derived from the occurrence of overcast and partly overcast sky conditions, and using an estimate of 0.3 for pD on clear days.

Determination of leaf chlorophyll, carbon and nitrogen contents

An estimate of leaf chlorophyll content was determined in situ for each leaf by Minolta SPAD 502 Chlorophyll Meter (Minolta Camera Co., Ltd, Osaka, Japan). SPAD readings were obtained in five to eight different leaf locations, and an average was stored. To calibrate the SPAD values, we selected for each species from each site 12 to 15 leaves of contrasting SPAD readings for destructive chlorophyll content estimations. For these leaves, a SPAD estimate was taken again, and leaf discs were punched from the immediate location of the SPAD measurement. Leaf chlorophylls were extracted from 5 mm-diameter leaf discs with dimethyl sulphoxide (DMSO) for 2.5 h at 65 °C as in Barnes et al. (1992). Absorptance of the extracts was determined with a spectrophotometer Spectronic 2000 (Bausch & Lomb, Rochester, NY, USA), and leaf chlorophyll concentrations were calculated according to Barnes et al. (1992). These data were used to derive non-linear relationships between leaf chlorophyll content and SPAD values. As observed previously (Richardson, Duigan & Berlyn 2002), best fits to the data were obtained by exponential regressions (Fig. 1). For each leaf, chlorophyll content was calculated from the SPAD value using species- and site-specific regressions (Fig. 1).


Figure 1. Calibration of Minolta Chlorophyll Meter (SPAD 502; Minolta Camera Co., Ltd, Osaka, Japan) readings for Ilex aquifolium (filled symbols) and Rhododendron ponticum (open symbols) leaves sampled at the Spanish (a) and Belgian (b) sites. Chlorophyll content was determined spectrophotometrically after extraction with dimethyl sulphoxide (DMSO). The data were fitted by non-linear regressions.

Download figure to PowerPoint

Foliar nitrogen (NM) and carbon (CM) contents were measured using a gas chromatograph after flash-combustion of the sample in oxygen by a NC 2100 elemental analyser (Carlo Erba Institute, Milan, Italy).

Chlorophyll fluorescence measurements

In the Belgian site, a pulse-amplitude modulated fluorometer (PAM-2000; Heinz Walz GmbH, Effeltrich, Germany) with leaf clip holder (model 2030-B) was used to estimate the in situ light-saturated rate of linear photosynthetic electron transport (Jmax, µmol m−2 s−1; Schreiber, Bilger & Neubauer 1994):

Jmax  =  0.5ΦPSIIΘQ(2)

where Q is the photosynthetically active quantum flux density, Θ is leaf absorptance, and ΦPSII is the effective quantum yield of photosystem II. ΦPSII is given as:

ΦPSII  =  (Fm′  −  F)/Fm′(3)

where F is the fluorescence yield in actinic light, and Fm′ is the maximum fluorescence yield of the light-adapted sample. For these estimates, we determined leaf absorptance from leaf chlorophyll content using an empirical equation of Evans (1993). Overall, Eqn 2 provides a reliable measure of foliar photosynthetic electron transport (Edwards & Baker 1993) that is only moderately affected by partitioning of absorbed light between the two photosystems (Laisk et al. 2001).

During the measurements, low natural background diffuse irradiance of 5–100 µmol m−2 s−1 was supplemented by the actinic irradiance from the internal halogen lamp of PAM-2000, and the measurements were conducted at Q values of 200–300, 700–800, 1200–1400, 1700–1800 µmol m−2 s−1. Starting with the lowest Q, pulses of white light of 8000 µmol m−2 s−1 were applied to determine Fm′. ΦPSII was computed online, and the measurements were continued at each light level until the steady-state quantum yield was achieved. For each leaf, Jmax was defined as the average of the three highest measurements. During the fluorescence measurements, average leaf temperature across individual leaves varied from 19.3 to 24.9 °C (mean ± SD = 21.5 ±  1.2 °C).

Statistical analyses

Because the relationships between Qint and leaf morphological and structural characteristics were generally non-linear (see insets in Fig. 2a & b for the sample fits), Qint was transformed before the statistical analysis using log10-transformation. Thus, linear regression analyses could be employed to explore the relationships between the studied foliar variables and Qint. Species and site differences were separated by covariance analyses (ancova). First, a separate slope ancova analysis was carried out to test for the slope differences. When differences in the slopes were statistically insignificant, the intercepts were compared by a common slope ancova analysis (Sokal & Rohlf 1995).


Figure 2. Effects of seasonal average integrated quantum flux density (Qint) on (a, b) leaf dry mass per unit area (MA), and (c, d) leaf thickness (T) in I. aquifolium (filled symbols) and R. ponticum (open symbols) at the Spanish (a, c) and the Belgian site (b, d). Error bars give ± SE. Log(Qint) versus MA and T relationships were fitted by linear regressions (Table 1). Insets in (a) and (b) demonstrate the relationships between non-transformed Qint and MA.

Download figure to PowerPoint

Total phenotypic variability in the samples and variability within and among plants were estimated by the coefficient of variation, namely the standard deviation (SD) divided by the mean of the characteristic (see Valladares et al. 2002a and references therein). Phenotypic plasticity was defined as the fraction of total phenotypic variability that was unambiguously related to environmental variability; in our case to the variability in Qint. By examination of the 25, 50 and 75 percentiles of Qint distribution, leaf samples were separated in four categories: deep shade (<3.5 mol m−2 d−1), shade (3.6–6.7 mol m−2 d−1), moderate shade (6.8–11 mol m−2 d−1), and open or partial shade (>11 mol m−2 d−1). Values for R. ponticum corresponding to Qint lower than 1.4 mol m−2 d−1, which were only obtained in the Belgian population, were discarded to keep equivalent ranges of environmental variation for the two populations. Phenotypic plasticity was estimated by both the coefficient of variation between the means of these four groups and a plasticity index calculated as the difference between the minimum and the maximum mean values among these four light levels divided by the maximum mean value (Valladares et al. 2000a, b). Only key foliar structural variables were included in the latter analysis.


  1. Top of page

Plastic morphological response to light

Lamina dry mass per unit area (MA) scaled positively with seasonal average integrated quantum flux density (Qint) in both I. aquifolium and R. ponticum and at both the Spanish and the Belgian site (Fig. 2a & b). This positive relationship resulted from changes in lamina thickness (T, Fig. 2c & d) and density (D, Fig. 3a & b) with Qint (MA = D · T). Similarly to D, lamina dry to fresh mass ratio was positively correlated with irradiance in both species (Fig. 3c & d).


Figure 3. Leaf density (a, b) and dry to fresh mass ratio (c, d) in relation to Qint in I. aquifolium (filled symbols) and R. ponticum (open symbols) in both sites. Data presentation as in Fig. 2.

Download figure to PowerPoint

Average projected area per leaf (AP) as well as the silhouette leaf area with its natural three-dimensional shape (AS) was independent of Qint in all cases (P > 0.1). However, the AS/AP ratio was negatively related to Qint in R. ponticum at the Spanish site (Fig. 4a), and in I. aquifolium at the Belgian site (Fig. 4b), indicating that the leaves were more curled and rolled at higher irradiance in these cases. Leaf perimeter to AP ratio was positively associated with Qint in I. aquifolium at the Belgian site (Fig. 4d). Thus, the leaves of this species became more spiny with increasing irradiance.


Figure 4. Leaf silhouette to projected area ratio (a, b) and perimeter to projected area ratio (c, d) in dependence on Qint in I. aquifolium (filled symbols) and R. ponticum (open symbols) in both sites. Silhouette area is the leaf horizontal projection with its natural three-dimensional shape, whereas the projected area was estimated after pressing the leaf to a flat section. Non-significant regressions (P > 0.05) are not drawn.

Download figure to PowerPoint

Petiole to total leaf dry mass ratio was negatively associated with Qint in R. ponticum in both sites (Fig. 5a & b), whereas in I. aquifolium, this variable increased with increasing Qint at the Spanish site (Fig. 5a), and was independent of Qint at the Belgian site (Fig. 5b).


Figure 5. Correlations between Qint and petiole to total leaf dry mass ratio in I. aquifolium (filled symbols) and R. ponticum (open symbols) in both sites. Data presentation as in Fig. 4.

Download figure to PowerPoint

Variation in foliar chemistry along the light gradients

Leaf nitrogen content per unit dry mass (NM) was independent of irradiance in both species at the Spanish site (Fig. 6a), whereas it was negatively related to Qint at the Belgian site (Fig. 6b). Increases in MA with irradiance (Fig. 2a & b) also led to a positive relationship between leaf nitrogen content per unit leaf area (NA) and irradiance (NA = MANM) with r2 varying from 0.56 to 0.81 (P < 0.001 for all) for different site and species combinations.


Figure 6. Leaf nitrogen (a, b), and chlorophyll contents (c, d) per unit dry mass, and chlorophyll to nitrogen ratio (e, f) relative to Qint in I. aquifolium (filled symbols) and R. ponticum (open symbols) in both sites. Data presentation as in Fig. 4.

Download figure to PowerPoint

Chlorophyll content per unit dry mass (CM; Fig. 6c & d) and chlorophyll to nitrogen ratio (Fig. 6e & f) increased with decreasing irradiance in all cases, indicating an enhanced investment of resources to improve light harvesting in low irradiance. However, positive effects of light on MA were larger than the negative effects on CM, resulting in positive correlations between Qint and chlorophyll content per unit area (CA; Table 1).

Table 1.  Effects of seasonal average integrated quantum flux density [Log(Qint, mol m−2 d−1] on foliar morphology and chemistry in Ilex aquifolium and Rhododendron ponticum: linear regression analyses a
Independent variableSpanish siteBelgian site
I. aquifoliumR. ponticumI. aquifoliumR. ponticum
  1. a The explained variances of the regressions (r2) are reported in Figs 2–6. For Chl/area versus log(Qint) relationships, the values of r2 were 0.70 (I. aquifolium, Spanish site), 0.68 (R. ponticum, Spanish site), 0.55 (I. aquifolium, Belgian site), 0.40 (R. ponticum, Belgian site).

  2. b *** P < 0.001; ** P < 0.01; * P < 0.05; nsP > 0.05.

Leaf dry mass per unit area (g m−2) 56.2*** 54.8*** 27.9*** 98.9*** 63.7*** 85.1*** 51.7*** 96.8***
Leaf thickness (mm)  0.345***  0.104***  0.244***  0.183***  0.365***  0.224***  0.237***  0.148***
Leaf density (g cm−3)  0.176***  0.0685***  0.176***  0.116***  0.182***  0.0734***  0.219***  0.161***
Leaf dry to fresh mass ratio (g g−1)  0.252***  0.0811***  0.245***  0.128***  0.283***  0.0913***  0.273***  0.151***
Leaf projected to silhouette area ratio  0.973*** −0.0122 ns  1.01*** −0.0325*  0.972*** −0.0559*  0.988*** −0.0046 ns
Leaf perimeter to projected area ratio (cm−1)  1.08***  0.0290 ns  0.975***  0.075 ns  1.21***  0.482*  0.881** −0.038 ns
Petiole to total leaf dry mass ratio (g g−1)  0.0544***  0.0127**  0.0541*** −0.0090**  0.0269***  0.0068 ns  0.0424*** −0.0104***
Leaf nitrogen content (%)  1.63***  0.117 ns  1.14*** −0.058 ns  2.10*** −0.420*  1.42*** −0.536***
Leaf chlorophyll content (µmol g−1)  6.80*** −0.871**  8.69*** −2.35***  9.17*** −3.26*** 10.3*** −4.97***
Leaf chlorophyll content (µmol m−2)b398***240***348***420***632***256***562***190***
Chl/N ratio (mmol mol−1)  5.66*** −0.900** 10.5*** −2.31***  6.19*** −1.23** 10.3*** −1.85***

Interspecific differences in plastic morphological response to light

The slope of the MA versus Qint relationship was significantly larger (P < 0.01) in R. ponticum than in I. aquifolium in the Spanish site (Fig. 2a; Table 1), but the slopes and intercepts (P > 0.4 for both comparisons) were not different between the species at the Belgian site (Fig. 2b). A larger slope in R. ponticum at the Spanish site resulted from a greater plasticity in both lamina thickness (Fig. 2c; P < 0.001) and density (Fig. 3a). At the Belgian site, the slope of leaf density versus Qint relationship was also larger in R. ponticum (Fig. 3b; Table 1; P < 0.005), but the slope of thickness versus Qint dependence (Fig. 2d; Table 1) was larger in I. aquifolium (P < 0.05). This led to similar plasticity in MA (T · D) in both species at the Belgian site. As a general pattern, leaf density was larger in R. ponticum than in I. aquifolium over most of the understorey light gradient in both sites (Fig. 3a & b). Greater plasticity in D was also paralleled by more extensive changes in lamina dry to fresh mass ratio in R. ponticum in both sites (Fig. 3c & d; P <  0.001 for the species difference at the Spanish, and P <  0.05 at the Belgian site).

Leaf silhouette to projected area ratio tended to be lower in I. aquifolium than in R. ponticum at both sites (Fig. 4a & b), indicating lower light-harvesting efficiency in I. aquifolium. Although the leaves of I. aquifolium were wider and shorter than those of R. ponticum, leaf perimeter to projected area ratio was not significantly different (P > 0.1) between the species at the Spanish site (Fig. 4c). At the Belgian site, this ratio was even larger over most of the irradiance range in I. aquifolium (Fig. 4d). These results provide indirect evidence of a large foliar investment in spines, that is the structural defence in I. aquifolium.

Foliar dry mass investment in petioles was larger in I. aquifolium at the Spanish site (Fig. 5a), whereas contrasting patterns were observed at the Belgian site (Fig. 5b).

Species differences in foliar nitrogen and chlorophyll contents

Leaf nitrogen content per unit dry mass was significantly larger at a common irradiance (P < 0.001) in I. aquifolium than in R. ponticum in both sites (Fig. 6a & b). This difference also resulted in a greater NA in I. aquifolium (P < 0.001, cf. Figs 2a & b and 6a & b).

Leaf chlorophyll content per unit dry mass increased more with decreasing Qint in R. ponticum than in I. aquifolium at the Spanish site (P < 0.05; Fig. 6c). The same pattern was also evident at the Belgian site (P = 0.06; Fig. 6d). At the Spanish site, chlorophyll content per unit area also increased more with increasing irradiance in R. ponticum (P < 0.001; Table 1), but not at the Belgian site (P > 0.3). Chlorophyll to nitrogen ratio was consistently larger in R. ponticum at both sites (Fig. 6e & f).

Species differences in photosynthetic capacity in relation to foliar structure and nitrogen

The light-saturated photosynthetic electron transport rate (Jmax) per unit area (Jmax; A) was positively associated with MA(Fig. 7a) and with NA (Fig. 7b) in both species at the Belgian site. The correlation between Jmax per unit dry mass (Jmax; M) and MA was not significant (Fig. 7c). Jmax; M was strongly related to NM, indicating that the positive relationship between Jmax; A and nitrogen content per unit area was determined both by changes in leaf nitrogen concentration and accumulation of photosynthetic biomass per unit area with MA.


Figure 7. Relationships of photosynthetic electron transport rate per unit area (a, b) and per unit dry mass (c, d) with leaf dry mass per unit area (a, c) and leaf nitrogen content per unit area (b) and dry mass (d) in I. aquifolium (filled symbols) and R. ponticum (open symbols). The measurements were conducted at the Belgian site. The data were fitted by linear regression.

Download figure to PowerPoint

The slopes of Jmax; A versus MA (P > 0.6; Fig. 7a), Jmax; A versus NA (P > 0.05; Fig. 7b) and Jmax; M versus NM (P > 0.8; Fig. 7d) relations were not significantly different between the species. However, the intercepts of Jmax; A versus NA (Fig. 7b) and Jmax; M versus NM (Fig. 7d) relations were significantly larger in R. ponticum than in I. aquifolium (P < 0.001 for both comparisons), demonstrating that the photosynthetic nitrogen use efficiency is higher in R. ponticum.

General trends and species–site interactions in phenotypic variability and plasticity

Although phenotypic variability among leaves of a given cluster (either whorls in R. ponticum or adjacent leaves in a stem or branch in I. aquifolium) was very low, variability among plants of the same species and site was quite high, accounting for the largest fraction of the total mean variability observed for eight foliar variables (Table 2). The Belgian population of R. ponticum exhibited significantly larger phenotypic variability than that of the Spanish one and than those of both the Belgian and the Spanish populations of I. aquifolium (Table 2). Mean phenotypic plasticity, defined as the fraction of total phenotypic variability unambiguously linked to variability in Qint, was significantly larger in the Belgian population of R. ponticum than in the Spanish population (Table 3).

Table 2.  Total, among, and within plant variability of eight foliar variables in Ilex aquifolium and Rhododendron ponticum at the two sites. Variability was estimated by the coefficient of variation (SD/mean)
Site and group Coefficient of variationSignificance
Leaf dry mass per unit areaLeaf thicknessLeaf densityLeaf dry to fresh mass ratioPetiole to total leaf dry massLeaf N content per dry massLeaf chlorophyll content per areaChl/N molar ratioMean
  1. Significant (P < 0.05, anova, Tukey test) differences among the means are indicated by a letter code; ns. no significant differences.

Total variability (n = 503 samples)
 I. aquifoliumSpain0.2820.1390.1690.1470.1600.1550.2340.1440.179a
 R. ponticumSpain0.3260.1840.2020.1540.1700.1830.2570.1760.206a, b
Variability among plants (n = 42 plants)
 I. aquifoliumSpain0.1820.0570.1200.1340.1760.0350.1130.0250.105a
 R. ponticumSpain0.1640.1100.0900.0790.0600.0300.0970.0910.090a
Variability within plants (n = 104 leaf clusters)
 I. aquifoliumSpain0.0680.0390.0610.030  0.066 0.053ns
Belgium0.0780.0520.0690.045  0.084 0.066ns
 R. ponticumSpain0.0630.0540.0610.028  0.064 0.054ns
Belgium0.0560.0410.0450.028  0.068 0.048ns
Table 3.  Phenotypic plasticity (i.e. variability related to differences in light availability) for eight foliar variables in Ilex aquifolium and Rhododendron ponticum at the two sites. Plasticity is estimated by both the coefficient of variation and by a plasticity index [(Maximum-minimum)/maximum], Valladares et al. (2000a). Only samples within equivalent ranges of light availability were considered
SpeciesSiteLeaf dry mass per unit areaLeaf thicknessLeaf densityLeaf dry to fresh mass ratioPetiole to total leaf dry massLeaf N content per dry massLeaf chlorophyll content per areaChl/N molar ratioMeanSignificance
  1. Significant (P < 0.05, anova, Tukey test) differences among the means are indicated by a letter code.

Coefficient of variation
 I. aquifoliumSpain0.2540.1060.1430.1410.1240.0840.2160.0820.144a, b
 R. ponticumSpain0.3260.1530.1890.1120.0660.0650.1870.1200.152a
Plasticity index
 I. aquifoliumSpain0.4180.2050.2670.2770.2240.1860.3230.1820.260a, b
 R. ponticumSpain0.5280.2980.3320.2300.1380.1380.3480.2120.278b

Ilex aquifolium was significantly more plastic at the Belgian than at the Spanish site with respect to increases in MA (P < 0.02 for slope differences, Table 1), leaf thickness (P < 0.005), and for decreases in CM (P < 0.001) with increasing Qint. The slopes were not different, but leaf dry to fresh mass ratio (P < 0.001) and CA (P < 0.001) were larger at a common Qint in this species at the Belgian site. Neither slopes nor intercepts of leaf density and chlorophyll to N ratio versus Qint relations of I. aquifolium differed between the sites. Thus, site differences in CM were primarily modified by site differences in NM versus Qint relations (Fig. 6a & b).

In R. ponticum, CM decreased significantly more (P < 0.005) and CA increased significantly less with increasing Qint at the Belgian site. Again, chlorophyll to N ratio was not different between the sites in this species (P > 0.4), indicating that site differences in foliar chlorophyll versus irradiance relations resulted from contrasting variation patterns in foliar N availability for chlorophyll formation. There were no other site differences in slopes in R. ponticum. However, MA, D and dry to fresh mass ratio were significantly larger at a common irradiance at the Belgian site (P < 0.001; Table 1), whereas T and petiole to total leaf mass ratio were larger at the Spanish site (P < 0.001; Table 1) in this species.


  1. Top of page

Plastic morphological responses to light

Both R. ponticum and I. aquifolium (Figs 2 & 3) exhibited a significant morphological plasticity to seasonal average integrated irradiance (Qint). Increases in leaf dry mass per unit area (MA) with Qint allow accumulation of photosynthetic compounds per unit leaf area, and provide the primary explanation for the scaling of leaf photosynthetic capacities with irradiance (Niinemets & Tenhunen 1997). In turn, decreases in MA lead to formation of larger foliar area, and greater plant light interception with common biomass investment in foliage, which is a relevant mechanism for a vast number of plant species with contrasting life strategies (Valladares, Skillman & Pearcy 2002).

In our study, changes in MA resulted from modifications in both thickness (T; Fig. 2c & d) and leaf density (D, Fig. 3a & b). An increase in D with light may either indicate a larger relative contribution of tightly packed palisade mesophyll at the expense of spongy parenchyma, and/or smaller mesophyll cells with thicker and more lignified cell walls and lower fraction of intercellular air space (Niinemets 1999). This is an important difference, because the first modification increases leaf photosynthetic capacity, whereas the second one decreases it (Niinemets 1999). Larger D generally also results in a higher leaf elastic modulus and allows lower leaf water potentials for a common water loss. In fact, parallel changes in D and leaf dry to fresh mass ratio with Qint support the hypothesis of a larger structural investment at higher irradiance (Garnier et al. 1999). Given that foliar water limitations often increase with increasing Qint (Niinemets et al. 1999), such modifications in foliar structure may provide an important way to cope with high light environments. Thus, greater D (Fig. 3a & b) and dry to fresh mass ratio (Fig. 3c & d) at high irradiance in R. ponticum than in I. aquifolium in both sites may partly explain the larger competitive potential of R. ponticum in open habitats. However, leaf morphology versus Qint relationships were curvilinear in all cases (Fig. 2a & b), hinting at general limitations associated with the use of high irradiances in both species. As a rule, MA versus irradiance relationships are linear in true sun species (Niinemets & Kull 1998).

Leaf density of both I. aquifolium and R. ponticum was relatively low, being comparable with D estimates for temperate species (0.3–0.5 g cm−3, Niinemets 2001). In Mediterranean sclerophylls, D is considerably larger (0.6–0.8 g cm−3, Niinemets 2001). Given that low density is associated with low elastic modulus, and accordingly with a lower ability for water extraction from drying soil with a common leaf water loss (Niinemets 2001), this may explain the low tolerance of low soil water availability in both species. Thus, limited tolerance of water shortage may be the primary reason of constrained dispersal of both species at the Spanish site. Distribution of many Tertiary relicts in Mediterranean climates is curbed by the high sensitivity of these species to water stress (Gulías et al. 2002).

The species difference in D was also paralleled by higher dry to fresh mass ratio in R. ponticum in both sites. Furthermore, both of these variables were larger at a common irradiance in R. ponticum at the Belgian site. Thus, this species may be more drought tolerant in the naturalized habitat than at the native site, possibly as a result of hybridization with other introduced Rhododendron species (Milne & Abbott 2000).

Foliage light-harvesting efficiency

Leaf silhouette to projected area was generally larger in R. ponticum than in I. aquifolium (Fig. 4a & b), in particular, in higher irradiance. This difference implies higher effective leaf area for light interception and greater light-harvesting efficiency in R. ponticum, further contributing to higher competitive potential of this species.

Leaf chlorophyll contents per unit dry mass (CM) increased with decreasing Qint in all situations (Fig. 6c & d). However, there was evidence of greater CM, and accordingly, greater light absorptance per unit leaf mass in lower irradiance in R. ponticum in both cases (Fig. 6c & d). Furthermore, nitrogen partitioning to chlorophyll and associated proteins of leaf light-harvesting apparatus was significantly larger in R. ponticum (Fig. 6e & f), demonstrating that this species achieved higher light-harvesting efficiency in limited irradiance with lower nitrogen investments in leaves.

Chlorophyll to N ratio differs little among deciduous broad-leaved temperate forest species with values varying from approximately 9 to 4 mmol mol−1 along the canopy irradiance gradient (Niinemets et al. 1998). These values are comparable with those in I. aquifolium in high and intermediate light, whereas the chlorophyll to N ratios were higher in R. ponticum than in co-existing broad-leaved deciduous trees. This may mean that inherently high nitrogen partitioning to light harvesting provides a further explanation for the invasiveness of this species.

According to previous studies, R. ponticum may exist in forest understoreys at less than 2% of incident irradiance (Cross 1975; Tabbush & Williamson 1987). However, it generally becomes less competitive at irradiances below 5–10% of full light (Tabbush & Williamson 1987), and its growth rates are the largest in the highest light treatments (Cross 1975). This may be linked to limited N availability for new foliage formation and increase in leaf chlorophyll concentrations at low irradiance. Therefore, it is important that the foliar nitrogen concentrations were higher in low Qint at the Belgian site (Fig. 6a & b), possibly because of enhanced N deposition rates. The hypothesis of N and chlorophyll controls on species dispersal to understorey patches with low irradiance is supported by the existence of leaves in both species at lower irradiances at the Belgian site: 0.43 mol m−2 d−1 (1.4% of above-ground irradiance) for R. ponticum and 0.53 mol m−2 d−1 (1.8%) for I. aquifolium, than at the Spanish site: 1.65 mol m−2 d−1 (4.1%) for R. ponticum and 1.39 mol m−2 d−1 (3.5%) for I. aquifolium (Fig. 2).

High N depositions at the Belgian site only enhanced foliar N concentrations in low irradiance (Fig. 6a & b). However, there is evidence of considerably higher leaf area indices of R. ponticum in the Belgian (approximately 1.3 m2 m−2 scaled up to the entire stand, Nadezhdina et al. 2003) than at the Spanish site (approximately 0.2 m2 m−2), and the overstorey trees of P. sylvestris at the Belgian site have high N concentrations on the order of 2% (Janssens et al. 1999). Thus, N availability is apparently higher at the Belgian site, but R. ponticum increased foliar area rather than N of single leaves in response to increased N loads. Enhanced foliar area formation is compatible with greater plant light interception, and with superior plant performance in forest understorey. Given that nitrogen concentration of R. ponticum leaves was lower (Fig. 6a & b), this species can form a larger foliar display with equal total foliar N content. This indicates that the invasive R. ponticum can more efficiently capitalize on enhanced nutrient availability than the co-existing native species I. aquifolium.

Photosynthetic nitrogen use efficiency

We observed higher foliar photosynthetic capacity at a common leaf nitrogen content in R. ponticum than in I. aquifolium (Fig. 7b & d). Thus, both species had a similar rate of net photosynthesis per unit area (Fig. 7a), but R. ponticum achieved this with lower N investments in leaves. Given that the specific activities of the enzymes participating in the dark and light reactions of photosynthesis are believed to be very conserved among C3 plants (Niinemets & Tenhunen 1997; Evans 1998; Evans & Poorter 2001), species differences in photosynthetic nitrogen use efficiency may directly result from inherently lower investments of foliar N in photosynthesis apparatus in I. aquifolium.

Species do significantly differ in N partitioning in photosynthetic apparatus (Evans & Poorter 2001; Le Roux et al. 2001). As our study suggests, high fractional N investments in photosynthetic machinery may be an important feature of invasive plant species. Similarly to Rhododendron, the photosynthetic nitrogen use efficiency is high and may vastly be enhanced by low site nutrient availability in evergreen dwarf-bamboo Sasa senanensis that is a strong invader in temperate broad-leaved Japanese forests (Lei & Koike 1998; Wang et al. 2001).

Evergreen leaves of Rhododendron may fix a large amount of nutrients and make them unavailable to other species (Monk, McGinty & Day 1985). Combined with the allelopathy and slow turnover of Rhododendron litter (Nilsen et al. 1999; Walker et al. 1999; Wright & Coleman 2000), the ecosystems dominated by Rhododendron are generally low in nutrients. This is similar in S. senanensis that also produces litter with a low decomposition rate due to the occurrence of allelopathic compounds (Eyin, Jayakumar & Pannirselvam 1989). Thus, inherently high photosynthetic nitrogen use efficiency may be a general trait of invasive species with ample production of toxic secondary compounds, effectively compensating for slow litter turnover rates. It is further important that increased N availability may significantly enhance the competitive potential of these species via enhanced foliage production, which allows them to shade out the competitors.

The role of phenotypic variability and plasticity in species invasiveness

Introduced species often hybridize with native species (Strefeler et al. 1996; Milne & Abbott 2000), leading to an increased genetic diversity in the alien plant populations. In fact, introduced populations of R. ponticum frequently contain a number of hybrids (Tabbush & Williamson 1987) and there is molecular genetic evidence of hybridization of R. ponticum with North-American relatives in the British Isles (Milne & Abbott 2000). The higher phenotypic variability of the alien, Belgian population of R. ponticum in comparison with the native, Spanish population seems to reflect differences in the genetic variability of each population, which is likely to be higher in the Belgian population than in the relict and receding Spanish one. The presence of hybrids and the increased genotypic variability could have some bearing on the invasive vigour of alien R. ponticum populations.

It has been hypothesized that enhanced phenotypic plasticity plays an important role in successful colonization of new habitats by alien species (Abbott 1992; Williams, Mack & Black 1995; Yamashita et al. 2000; Durand & Goldstein 2001). In agreement with this hypothesis, our results indicate enhanced plasticity in response to light in R. ponticum, especially in the invasive population (Table 3). The relatively high phenotypic plasticity of R. ponticum, a shade-tolerant species, does not agree with the general trend of low plasticity in shade-tolerant species (Valladares et al. 2000b). However, a distinction must be made between plasticity in morphological versus physiological variables as discussed in Valladares et al. (2000b). It has been shown that shade-tolerant understorey species have low photosynthetic rates and low photosynthetic plasticity to irradiance (Valladares et al. 2000b; Yamashita, Koike & Ishida 2002), whereas the photosynthetic plasticity is particularly large in high light species (Valladares et al. 2000b). The reverse seems to be true for morphological and light capture variables, at least in species like the European beech (Valladares et al. 2002b). A physiologically plastic response to high light has been interpreted as a mechanism of invasive species to be more efficient at utilizing light than native species in high light environments associated with high levels of disturbance (Yamashita et al. 2000; Durand & Goldstein 2001; Yamashita et al. 2002). The relatively low plasticity of I. aquifolium, which agrees with the low photosynthetic plasticity in response to changes in growth irradiance observed by Schmidt, Batic & Pfanz (2000) in this species, suggest a conservative use of resources (Valladares et al. 2000a). This contrasts with the more opportunistic resource use of the invasive populations of R. ponticum. Yet, studies also demonstrate that nitrogen availability may be the critical factor limiting photosynthetic acclimation (Ferrar & Osmond 1986). In our study, the rate of photosynthetic electron transport varied along the light gradient by a factor of three in both species at the Belgian site that has high N depositions, demonstrating similar photosynthetic plasticity in both species (Fig. 7). However, in comparison with the native I. aquifolium, the alien population of R. ponticum had higher nitrogen investments in light harvesting (Fig. 6e & f) and in photosynthetic machinery (Fig. 7), allowing it to produce more foliar area with equal photosynthetic and light-harvesting characteristics.

As suggested by Dukes & Mooney (1999), invasive plant species possess a set of traits that allow them to capitalize on increasing nitrogen availability. As our work shows, enhanced nitrogen investment in light-harvesting and photosynthetic compounds combined with high rates of leaf area production could partly explain the invasive nature of R. ponticum. Paralleling the increase in nitrogen deposition rates, there has been a general increase in the coverage of Rhododendron species in several places in the world (Nilsen et al. 1999; Esen 2000), providing indirect evidence of the role of nitrogen in the invasion potential of Rhododendron species. The high nitrogen use efficiency of R. ponticum together with its relatively high plastic response to light, despite its shade tolerance, and its capacity to hybridize and enhance its genetic and phenotypic variability, are the new features to be added to our understanding of the success of R. ponticum as an invading species in many temperate, humid and subhumid regions of Europe.


  1. Top of page

We thank the Belgian Federal Office for Scientific, Technical and Cultural Affairs (OSTC; research fellowship to Ü. Niinemets in frames of a project for the encouragement of the scientific and technical co-operation with Central and Eastern Europe), the Spanish Council for Scientific Research (CSIC) and the Estonian Academy of Sciences (collaborative project between research institutions of CSIC and research institutions in Estonia), the Estonian Science Foundation (grant 4584), and the German Academic Exchange Service (equipment grant to Ü.N) for funding this work. Scientific advise, logistic help and companionship in the Spanish field site from Teodoro Marañón (IRNA-CSIC), Juan Arroyo (Universidad de Sevilla) and Miguel Angel Zavala (Universidad de Alcalá), all belonging to the thematic network GLOBIMED (Spanish MCYT), is gratefully recognized. We appreciate ingenious technical assistance of Wendy Hendrickx, Frosina Uzunova and Daniela Brites, and thank the Institute for Forestry and Game Management (Geraardsbergen), and the Ministry of the Flemish Community (Brussels) for access to the forest site.


  1. Top of page
  • Abbott R.J. (1992) Plant invasions, interspecific hybridization and the evolution of new plant taxa. Trends in Ecology and Evolution 7, 401405.
  • Alados I., Foyo-Moreno I. & Alados-Arboledas L. (1996) Photosynthetically active radiation: measurements and modelling. Agricultural and Forest Meteorology 78, 121131.
  • Alados-Arboledas L., Olmo F.J., Alados I. & Pérez M. (2000) Parametric models to estimate photosynthetically active radiation in Spain. Agricultural and Forest Meteorology 101, 187201.
  • Aparicio A., Porras C. & Ceballos G. (2001) Bosques-Isla de la Provincia de Cádiz, Consejería de Medio Ambiente de la Junta de Andalucía, Sevilla, Spain (in Spanish).
  • Barnes J.D., Balaguer L., Manrique E., Elvira S. & Davison A.W. (1992) A reappraisal of the use of DMSO for the extraction and determination of chlorophylls a and b in lichens and higher plants. Environmental and Experimental Botany 32, 85100.
  • Benfield E.F. & Webster J.R. (1991) Effects of forest disturbance on leaf breakdown in southern Appalachian streams. Limnologie 24, 16781690.
  • Blanca G., Cabezudo B., Hernández-Bermejo J.E., Herrera C.M., Muñoz J. & Valdés B. (200001) El Libro Rojo de las Especies de Flora Amenazada en Andalucía, Vol. I–II. Consejería de Medio Ambiente de la Junta de Andalucía, Sevilla, Spain (in Spanish).
  • Brown J.M.B. (1953) The Rhododendron problem in the woodlands of southern England. Quarterly Journal of Forestry 47, 239253.
  • Cross J.R. (1975) Biological flora of the British Isles. List Br. Vasc. Pl. (1958), 345.1: Rhododendron ponticum L. Journal of Ecology 63, 345364.
  • Cross J.R. (1981) The establishment of Rhododendron ponticum in the Killarney oakwoods, S. W. Ireland. Journal of Ecology 69, 807824.
  • Davis M.A., Grime J.P. & Thompson K. (2000) Fluctuating resources in plant communities: a general theory of invasibility. Journal of Ecology 88, 528534.
  • Dukes J.S. & Mooney H.A. (1999) Does global change increase the success of biological invaders? Trends in Ecology and Evolution 14, 135139.
  • Durand L.Z. & Goldstein G. (2001) Photosynthesis, photoinhibition, and nitrogen use efficiency in native and invasive tree ferns in Hawaii. Oecologia 126, 345354.
  • Edwards G.E. & Baker N.R. (1993) Can CO2 assimilation in maize leaves be predicted accurately from chlorophyll fluorescence analysis? Photosynthesis Research 37, 89102.
  • Esen D. (2000) Ecology and control of rhododendron (Rhododendron ponticum L.) in Turkish eastern beech (Fagus orientalis Lipsky) forests. PhD Dissertation, Virginia Polytechnic Institute and State University, Blacksburg, VA, SA.
  • Evans J.R. (1993) Photosynthetic acclimation and nitrogen partitioning within a lucerne canopy. II. Stability through time and comparison with a theoretical optimum. Australian Journal of Plant Physiology 20, 6982.
  • Evans J.R. (1998) Photosynthetic characteristics of fast- and slow-growing species. In Inherent Variation in Plant Growth. Physiological Mechanisms and Ecological Consequences (eds H.Lambers, H.Poorter & M.M.I.Van Vuuren), pp. 101119. Backhuys Publishers, Leiden, The Netherlands.
  • Evans J.R. & Poorter H. (2001) Photosynthetic acclimation of plants to growth irradiance: the relative importance of specific leaf area and nitrogen partitioning in maximizing carbon gain. Plant, Cell and Environment 24, 755767.
  • Eyin M., Jayakumar M. & Pannirselvam S. (1989) Allelopathic effect of bamboo leaf extract on the seedling of groundnut. Tropical Ecology 30, 138141.
  • Ferrar P.J. & Osmond C.B. (1986) Nitrogen supply as a factor influencing photoinhibition and photosynthetic acclimation after transfer of shade-grown Solanum dulcamara to bright light. Planta 168, 563570.
  • Garnier E., Salager J.-L., Laurent G. & Sonié L. (1999) Relationships between photosynthesis, nitrogen and leaf structure in 14 grass species and their dependence on the basis of expression. The New Phytologist 143, 119129.
  • Gond V., De Pury D.G.G., Veroustraete F. & Ceulemans R. (1999) Seasonal variations in leaf area index, leaf chlorophyll, and water content: scaling-up to estimate fAPAR and carbon balance in a multilayer, multispecies temperate forest. Tree Physiology 19, 673679.
  • Gucci R., Lombardini L. & Tattini M. (1997) Analysis of leaf water relations in leaves of two olive (Olea europaea) cultivars differing in tolerance to salinity. Tree Physiology 17, 1321.
  • Gulías J., Flexas J., Abadía A. & Medrano H. (2002) Photosynthetic responses to water deficit in six Mediterranean sclerophyll species: possible factors explaining the declining distribution of Rhamnus ludovici-salvatoris, an endemic Balearic species. Tree Physiology 22, 687697.
  • Hampe A. & Arroyo J. (2002) Recruitment and regeneration in populations of an endangered South Iberian Tertiary relict tree. Biological Conservation 107, 263271.
  • Huxman T.E. & Smith S.D. (2001) Photosynthesis in an invasive grass and native forb at elevated CO2 during an El Niño year in the Mojave Desert. Oecologia 128, 193201.
  • Janssens I.A., Sampson D.A., Čermák J., Meiresonne L., Riguzzi F., Overloop S. & Ceulemans R. (1999) Above- and belowground phytomass and carbon storage in a Belgian Scots pine stand. Annals of Forest Science 56, 8190.
  • Jela J.L. (1997) Potential British lepidopteran fauna on introduced Rhododendron ponticum. Holarctic Lepidoptera 4, 5961.
  • Jela J.L. & Lawton J.H. (1997) Insect herbivore loads on native and introduced plants: a preliminary study. Entomologia Experimentalis et Applicata 85, 275279.
  • Laisk A., Oja V., Rasulov B., Rämma H., Eichelmann H., Kasparova I., Pettai H., Padu E. & Vapaavuori E. (2001) A computer-operated routine of gas exchange and optical measurements to diagnose photosynthetic apparatus in leaves. Plant, Cell and Environment 25, 923943.
  • Le Roux X., Walcroft A.S., Daudet F.A., Sinoquet H., Chaves M.M., Rodrigues A. & Osorio L. (2001) Photosynthetic light acclimation in peach leaves: importance of changes in mass: area ratio, nitrogen concentration, and leaf nitrogen partitioning. Tree Physiology 21, 377386.
  • Lei T.T. & Koike T. (1998) Functional leaf phenotypes for shaded and open environments of a dominant dwarf bamboo (Sasa senanensis) in Northern Japan. International Journal of Plant Sciences 159, 812820.
  • Madueño J.M.M. (1991) Cartografía y Estadística de Usos y Coberturas Vegetales del Suelo en Andalucía. Evolución 1976–91, Consejería de Medio Ambiente de la Junta de Andalucía, Sevilla, Spain (in Spanish).
  • Milne R.I. & Abbott R.J. (2000) Origin and evolution of invasive naturalized material of Rhododendron ponticum L. in the British Isles. Molecular Ecology 9, 541556.
  • Monk C.D., McGinty D.T. & Day F.P. Jr (1985) The ecological importance of Kalmia latifolia and Rhododendron maximum in the deciduous forest of the Southern Appalachians. Journal of the Torrey Botanical Society 112, 187193.
  • Mooney H.A. (1999) Life out of bounds: bioinvasion in a borderless world. Nature 397, 665666.
  • Nadezhdina N., Tatarinov F. & Ceulemans R. (2003) Leaf area distribution and crown structure of Rhododendron understory in a coniferous. forest: description and scaling-up. Forest Ecology and Management in press.
  • Neirynck J., Roskams P. & Lust N. (1999) Resultaten van 6 jaar monitoring van de chemische samenstelling van neerslag en bodemwater in het level II-proefvlak te Brasschaat. Mededelingen van het Instituut voor Bosbouw en Wildbeheer 1, 5976. (in Dutch).
  • Neirynck J., Van Ranst E., Roskams P. & Lust N. (2002) Impact of decreasing throughfall depositions on soil solution chemistry at coniferous monitoring sites in northern Belgium. Forest Ecology and Management 160, 127142.
  • Niinemets Ü. (1999) Research review. Components of leaf dry mass per area – thickness and density – alter leaf photosynthetic capacity in reverse directions in woody plants. The New Phytologist 144, 3547.
  • Niinemets Ü. (2001) Climatic controls of leaf dry mass per area, density, and thickness in trees and shrubs at the global scale. Ecology 82, 453469.
  • Niinemets Ü. & Kull O. (1998) Stoichiometry of foliar carbon constituents varies along light gradients in temperate woody canopies: implications for foliage morphological plasticity. Tree Physiology 18, 467479.
  • Niinemets Ü. & Tenhunen J.D. (1997) A model separating leaf structural and physiological effects on carbon gain along light gradients for the shade-tolerant species Acer saccharum. Plant, Cell and Environment 20, 845866.
  • Niinemets Ü., Bilger W., Kull O. & Tenhunen J.D. (1998) Acclimation to high irradiance in temperate deciduous trees in the field: changes in xanthophyll cycle pool size and in photosynthetic capacity along a canopy light gradient. Plant, Cell and Environment 21, 12051218.
  • Niinemets Ü., Sõber A., Kull O., Hartung W. & Tenhunen J.D. (1999) Apparent controls on leaf conductance by soil water availability and via light-acclimation of foliage structural and physiological properties in a mixed deciduous, temperate forest. International Journal of Plant Sciences 160, 707721.
  • Nilsen E.T., Walker J.F., Miller O.K., Semones S.W., Lei T.T. & Clinton B.D. (1999) Inhibition of seedling survival under Rhododendron maximum (Ericaceae): could allelopathy be a cause? American Journal of Botany 86, 15971605.
  • Obeso J.R. (1997) Costs of reproduction in Ilex aquifolium: effects at tree, branch and leaf levels. Journal of Ecology 85, 159166.
  • Ojeda F., Marañón T. & Arroyo J. (2000) Plant diversity patterns in the Aljibe Mountains (S-Spain): a comprehensive account. Biodiversity and Conservation 9, 13231343.
  • Palamarev E. (1989) Paleobotanical evidence of the Tertiary history and origin of the Mediterranean sclerophyll dendroflora. Plant Systematics and Evolution 162, 93107.
  • Reddy P.S., Jamil K., Madhusudhan P., Anjani G. & Das B. (2001) Antibacterial activity of isolates from Piper longum and Taxus baccata. Pharmaceutical Biology 39, 236238.
  • Richardson A.D., Duigan S.P. & Berlyn G.P. (2002) An evaluation of noninvasive methods to estimate foliar chlorophyll content. The New Phytologist 153, 185194.
  • Roskams P., Sioen G. & Overloop S. (1997) Meetnet voor de Intensieve Monitoring van het Bosecosysteem in het Vlaamse Gewest. Resultaten 1991–92. Ministerie van de Vlaamse Gemeenschap, Instituut voor Bosbouw en Wildbeheer, Geraardsbergen, The Netherlands (in Dutch).
  • Rotherham I.D. (1983) The ecology of Rhododendron ponticum with special reference to its competitive and invasive capabilities. PhD Dissertation, University of Sheffield, Sheffield, UK.
  • Schmidt J., Batic F. & Pfanz H. (2000) Photosynthetic performance of leaves and twigs of evergreen holly (Ilex aquifolium L.). Phyton – Annales Rei Botanicae 40, 179190.
  • Schreiber U., Bilger W. & Neubauer C. (1994) Chlorophyll fluorescence as a nonintrusive indicator for rapid assessment of in vivo photosynthesis. In Ecophysiology of Photosynthesis (eds E.-D.Schulze & M.M.Caldwell) Ecological Studies 100, pp. 4970. Springer Verlag, Berlin, Germany.
  • Secretaría General Técnica, Ministerio de Medio Ambiente (2001) Calendario Meteorológico. Serie Monografías, Ministerio de Medio Ambiente, Dirección General del Instituto Nacional de Meteorología, Madrid, Spain (in Spanish).
  • Sinamba Difusión. (1998) La Información Ambiental de Andalucía. Consejeria de Medio Ambiente de la Junta de Andalucía, Sevilla, Spain (in Spanish).
  • Sokal R.R. & Rohlf F.J. (1995) Biometry. The Principles and Practice of Statistics in Biology Research, 3rd edn. W.H. Freeman, New York, USA.
  • Strefeler M.S., Darmo E., Becker R.L. & Katovich R.J. (1996) Isoenzyme characterization of genetic diversity in Minnesota populations of purple loosestrife Lythrum salicaria (Lythraceae). American Journal of Botany 83, 265273.
  • Tabbush P.M. & Williamson D.R. (1987) Rhododendron ponticum as a Forest Weed. Forestry Commission Bulletin 73. Her Majesty's Stationery Office, London, UK.
  • Valladares F., Balaguer L., Martínez-Ferri E., Perez-Corona E. & Manrique E. (2002a) Plasticity, instability and canalization: is the phenotypic variation in seedlings of sclerophyll oaks consistent with the environmental unpredictability of Mediterranean ecosystems? The New Phytologist 156, 457467.
  • Valladares F., Chico J.M., Aranda I., Balaguer L., Dizengremel P., Manrique E. & Dreyer E. (2002b) The greater seedling high-light tolerance of Quercus robur over Fagus sylvatica is linked to a greater physiological plasticity. Trees: Structure and Function 16, 395403.
  • Valladares F., Ferri-Martínez E., Balaguer L., Perez-Corona E. & Manrique E. (2000a) Low leaf-level response to light and nutrients in Mediterranean evergreen oaks: a conservative resource-use strategy? The New Phytologist 148, 7991.
  • Valladares F., Wright S.J., Lasso E., Kitajima K. & Pearcy R.W. (2000b) Plastic phenotypic response to light of 16 congeneric shrubs from a Panamanian rainforest. Ecology 81, 19251936.
  • Valladares F., Skillman J.B. & Pearcy R.W. (2002) Convergence in light capture efficiencies among tropical forest understory plants with contrasting crown architectures: a case of morphological compensation. American Journal of Botany 89, 12751284.
  • Van Ranst E., De Coninck F., Roskams P. & Vindevogel N. (2002) Acid-neutralizing capacity of forest floor and mineral topsoil in Flemish forests (North Belgium). Forest Ecology and Management 166, 4553.
  • Walker J.F., Miller O.K. Jr, Lei T., Semones S., Nilsen E. & Clinton B.D. (1999) Suppression of ectomycorrhizae of canopy tree seedlings in Rhododendron maximum L. (Ericaceae) thickets in the southern Appalachians. Mycorrhiza 9, 4956.
  • Wang W., Kazama M., Kitaoka S., Osaki M. & Koike T. (2001) Photosynthetic characteristics of Sasa senanensis grown under low nitrogen, potassium and phosphorous nutrient conditions. Bamboo Journal 18, 2336.
  • Williams D.G., Mack R.N. & Black R.A. (1995) Ecophysiology of introduced Pennisetum setaceum on Hawaii – the role of phenotypic plasticity. Ecology 76, 15691580.
  • Wright C.J. & Coleman D.C. (2000) Cross-site comparison of soil microbial biomass, soil nutrient status, and nematode trophic groups. Pedobiologia 44, 223.
  • Yamashita N., Ishida A., Kushima H. & Tanaka N. (2000) Acclimation to sudden increase in light favoring an invasive over native trees in subtropical islands, Japan. Oecologia 125, 412419.
  • Yamashita N., Koike N. & Ishida A. (2002) Leaf ontogenetic dependence of light acclimation in invasive and native subtropical trees of different successional status. Plant, Cell and Environment 25, 13411356.

Received 30 October 2002; received in revised form 7 January 2003; accepted for publication 7 January 2003